[hheir]

From my understanding, the symmetrical shape  would have the two dipoles cancel out on each of the OH ends. Similar to the way they do in CH2Cl. I am reading online that ethylene glycol is polar and CH2Cl is, which is confusing me. Why does that rule about dipoles cancelling out not apply to ethylene glycol?

[wildfyr]

The polarity of the hydroxy groups greatly outweighs the symmetry. Symmetry in polarity mostly matters around a single atom, like when comparing CCl₄ vs CH₂Cl₂. In reality the ethylene glycol is assuming many different steric configurations per microsecond, many of which are not symmetrical

[Babcock_Hall]

When one assesses overall polarity, there is more to the story than just dipole-dipole forces.  What are intermolecular forces do you believe are present?

[Enthalpy]

And even if the molecule were permanently symmetric, locally it still has strong dipoles, which suffices to interact with other molecules.

[Babcock_Hall]

With respect to CH₂Cl₂, it is important to realize that the shape of dichloromethane is roughly tetrahedral; therefore, there will be a net dipole.

[Vidya]

CH2Cl2 is not symmetrical in all directions and it is also not perfect tetrahedral as all atoms are not identical.
In CCl4 all atoms are identical and this gives it a perfect tetrahedral shape.

[Vidya]

From my understanding, the symmetrical shape  would have the two dipoles cancel out on each of the OH ends. Similar to the way they do in CH2Cl. I am reading online that ethylene glycol is polar and CH2Cl is, which is confusing me. Why does that rule about dipoles cancelling out not apply to ethylene glycol?

In ethylene glycol each carbon is sp3 hybridized and tetrahedral shape.
If all atoms are identical  only then it is considered as perfectly symmetrical tetrahedron and will be non polar.

[Enthalpy]

In ethylene glycol [...] If all atoms are identical  only then it is considered as perfectly symmetrical tetrahedron and will be non polar. [...]

The central C-C bond rotates. Even if it didn't, local polarisation would remain, enough to interact with polar molecules.

[Corribus]

Defining what one means by being "polar" is important, but I think most would agree that being polar means having a permanent dipole. Having local interactions with polar molecules and having a permanent dipole are two different things. Even hexane has transient dipoles and induced dipoles that permit interactions with other nearby molecules, so saying a molecule can interact with polar molecules doesn't mean a whole lot.

Strictly speaking, if ethylene glycol is drawn in the trans configuration, it is nonpolar. However due to the stabilization from intramolecular hydrogen bond formation, the gauche (cis, if you want) configuration is more stable, with a rotation barrier around the C-C bond of ~10 kcal/mol even in vapor phase (again, due to hydrogen bond the rotational barrier is pretty big). (see. Buckley and Gigere, Canadian J. Chem. 1967) The gauche configuration has a permanent dipole moment - hence most molecules in the ensemble at room temperature have a permanent dipole moment and are therefore "polar".

There's a good lesson here, though: even a "pure" sample of something isn't monolithic or static. At any given time, there are a distribution of structures (conformations). So, even if you would tend to draw a molecule in a symmetric "nonpolar" way, there will always be some molecules that are distorted in such a way to be polar due to vibrations, rotations, and so forth. The macroscopic properties are a function of the ensemble average.